《计算机应用研究》|Application Research of Computers

基于潜在标签挖掘和细粒度偏好的个性化标签推荐

Personalized tag recommendation based on potential tag mining and fine-grained preference

免费全文下载 (已被下载 次)  
获取PDF全文
作者 李红梅,刁兴春,曹建军,张磊,冯钦
机构 1.陆军工程大学,南京 210007;2.国防科技大学 第六十三研究所,南京 210007
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2020)01-007-0034-06
DOI 10.19734/j.issn.1001-3695.2018.05.0498
摘要 为进一步提高个性化标签推荐性能,针对标签数据的稀疏性以及传统方法忽略隐藏在用户和项目上下文中潜在标签的缺陷,提出一种基于潜在标签挖掘和细粒度偏好的个性化标签推荐方法。首先,提出利用用户和项目的上下文信息从大量未观测标签中挖掘用户可能感兴趣的少量潜在标签,将标签重新划分为正类标签、潜在标签和负类标签三类,进而构建〈用户,项目〉对标签的细粒度偏好关系,在缓解标签稀疏性的同时,提高对标签偏好关系的表达能力;然后,基于贝叶斯个性化排序优化框架对细粒度偏好关系进行建模,并结合成对交互张量分解对偏好值进行预测,构建细粒度的个性化标签推荐模型并提出优化算法。对比实验表明,提出的方法在保证较快收敛速度的前提下,有效地提高了个性化标签的推荐准确性。
关键词 个性化标签推荐; 潜在标签挖掘; 贝叶斯个性化排序; 成对交互张量分解
基金项目 国家自然科学基金资助项目
中国博士后科学基金资助项目
本文URL http://www.arocmag.com/article/01-2020-01-007.html
英文标题 Personalized tag recommendation based on potential tag mining and fine-grained preference
作者英文名 Li Hongmei, Diao Xingchun, Cao Jianjun, Zhang Lei, Feng Qin
机构英文名 1.Army Engineering University,Nanjing 210007,China;2.the 63rd Research Institute,National University of Defense Technology,Nanjing 210007,China
英文摘要 To further improve the performance of personalized tag recommendation, this paper argued that traditional methods ignore the potential and informative tags hidden in the context of users and items. Aimed at this, this paper proposed a novel personalized tag recommendation method BPR-PITF-P based on potential tag mining and fine-grained preference. Firstly, BPR-PITF-P leveraged the context information of both users and got to mine potential and useful tags, and got three kinds of tags: positive tags, potential tags, and negative tags. Based on the above, it translated the traditional pairwise preference into fine-grained preference relationship among user-item post and tags. This kind of treatment helped alleviate the sparse problem of tagging data. Second, combined with pairwise interaction tensor factorization method to predict preference value, BPR-PITF-P modeled the preference relationship based on the optimization criteria of Bayesian personalized ranking, and developed a personalized tag recommendation model followed by optimization algorithm. The comparison results show that this proposed method could improve tag recommendation performance in the premise of guarantee convergence speed.
英文关键词 personalized tag recommendation; potential tag mining; Bayesian personalized ranking; pairwise interaction tensor factorization
参考文献 查看稿件参考文献
 
收稿日期 2018/5/8
修回日期 2018/6/25
页码 34-39
中图分类号 TP301.6
文献标志码 A